Megan McArdle in a post holding bankers blameless in the problems beleaguering the financial sector observes:
There are two basic narratives of what happened. The first is that bankers had bad incentives: they took massive risks because the profits were so good in the up years that it was worth the risk of the bad, or because they could pass the risks onto some other sucker, or they thought Uncle Sugar would bail them out. The other narrative is that bankers had bad information: they didn’t understand the risks they were taking.
I’ve always preferred narrative B, because Narrative A doesn’t make much sense. The CEOs of big banks lost vast sums of money, and their jobs, most of their social status, and so forth. They held onto the worst tranches of their securities, which implies they didn’t know how badly they were going to blow up. Etc.
I find it vastly more plausible, if not so comforting, to believe that systems can occasionally produce bad results even if the incentives basically point in the right direction. The FICO score revolution was valuable, but we took it too far. The money sloshing around US markets disguised the problems, because people who got into trouble tapped their home equity, or in a pinch, sold the house at a tidy profit. Everyone from borrowers to regulators was getting the same bad signal, that their behavior was much less risky than it actually was.
Arguendo, let’s assume that Megan is right. What are the regulatory implications of a finding that bad information caused the problems with the financial sector?
The first, obvious conclusion would be that, rather than finding that there were no villains in the financial crisis as Megan would have it, that the ratings agencies were the villains. It was their job to evaluate risk. Their corporate failure is manifest. They may have been incapable of providing reliable information about risk but they kept right on taking in money anyway. Shouldn’t there be some restrictions on selling a service you’re incapable of providing?
The ratings agencies are the beneficiaries of enormous subsidies from the federal government. If the meltdown in the financial sector was caused by a failure of information and the government-anointed companies in the business of providing information are either constitutionally incapable of providing that service or were overwhelmed by the conflict of interest at the heart of their business, shouldn’t they be getting a lot more attention than they’ve been receiving?
Update
Here’s what the SEC is asking of the ratings agencies now. In my view it’s wholly inadequate, considering the magnitude of the situation they fomented.
I’m not sure Megan is holding bankers blameless, but simply observing that they got their just rewards for takings risks they did not fully perceive. As much as it pains me to “defend” investment bankers, whom I generally detest, they are not stupid, and they are not intentionally self destructive. This caught them by surprise.
I think you hit a nail squarely on the head re: the rating agencies. They are the CFA’s who are supposed to pull apart these complex securities issues and raise the flag in case of perceived trouble. Let me amplify a crucial point you raise: the ratings agencies put out opinions FOR A FEE. Period.
It is widely claimed that the bankers did same. But they also PUT THEIR PERSONAL CAPITAL AT RISK. Way big difference. And many were devastated. That’s as it should be, but it is a bright line distinction between the scoring judges and the players.
@Drew
‘As much as it pains me to “defend†investment bankers, whom I generally detest, they are not stupid, and they are not intentionally self destructive. This caught them by surprise.’
I think they were bewitched by mathematics. See, Recipe for Disaster: The Formula that Killed Wall Street (http://www.wired.com/techbiz/it/magazine/17-03/wp_quant?currentPage=all)
Preferring pleasure over pain is a human characteristic, human nature, neither good nor bad. When that predisposition is dysregulated we call it “greed”. Greed is a vice. It is bad. The bankers were greedy. That they shared the pain does not absolve them for the harm their greed has done to others who were, indeed, blameless.
Being self-destructive is another human characteristic. It’s called by various names, “original sin”, “the death wish”, etc. but it’s universal. As St. Paul said “The good that I would do I do not do, and the evil that I would not do I do”. Investment bankers aren’t immune.
However, that being said clearly the ratings agencies were not only greedy but actually dishonest. Clearly.
We can’t outlaw greed and we would be imprudent to try. However, we can require that the ratings agencies have a lot more skin in the game. I suggest that they be compelled to put up bonds.
I guess I’d have to see the ratings in question, but as a non-consumer of ratings, I’ve always assumed that ratings were comparative risk assessments. A triple-A bond is less at risk of default than a double-A bond.
Like most difficult things in life, it’s probably a little bit of this and a dab of that. I agree with McArdle’s main contention that the focus on compensation is wasted. And I might even go so far as to say that attempts to align compensation with public interest strike me as more akin to socialism than most things being labeled thus these days.
My contention is that if there are financial instruments too complicated for the consumer to understand, rating agencies to rate and for the regulatory system to monitor, it should be outlawed. I believe the indirect effect of simplification would likely be lowered compensation, but who knows?
PD, the problem right now is that whether a particular instrument it too complicated for a ratings agency to rate can only be determined after the fact. The problem is that they’re rating things they don’t have the ability to rate not that they’re not rating them at all.
I can only see two regulatory alternatives for that. Either incentivize the ratings agencies to exert more care by requiring surety bonds of them or introduce the financial equivalent of the FDA for new financial instruments. Such an agency would have all of the conflicts of interest of the current FDA several orders of magnitude over because of the sums involved. I don’t see that as being a workable solution.
Sam –
I think that’s right. And the mathematics were incomplete.
Dave –
Pursuit of self interest is an activity engaged in by the vast majority of people every day, and bound to result in conflicts, errors and harmed individuals. However, the ability to discern from afar garden variety self interest from “bad,” greedy behavior is a supernatural talent with which I am not familiar. Especially in light of the risks that were not fully understood.
It seems to me the question to ask is whether or not our social and business processes are constructed such that participant’s actions have consequences. In the case of the raters, the answer was no. In the case of the bankers the answer was yes. That doesn’t absolve them, nor was it sufficient to prevent their actions. But most got there heads handed to them. I don’t know what else to do. The world is not perfect or riskless. And regulatory solutions historically are weak reeds to lean upon.
I’m reminded of the housing debate, where it became fashionable to lay the blame at the feet of mortgage brokers and lenders. Fine and well. But the easy credit regulators and the mortgage seekers were every bit as much to blame. Every bit as much. Just no political future in calling it what it was.
I recall posting many months ago – with obviously no resonance – that the real source of the financial meltdown was a pervasive and longstanding culture of easy credit and quick profit. That culture extended from Wall Street, to social policy driven regulators, to housing speculators, to vote seeking politicians, to monetary authorities to everyman. Blood on everyones hands.
I understand the temptation to find an easy whipping boy, like “greedy” bankers. But that pursuit only guarantees that the same mistakes will be made in the future. Its a scalp, not a solution.
I’d prefer to find the structural flaws, like the rating agency set of incentives, and fix them. Hence the bonding suggestion may have merit.
But I ask any of the legal types: what, then, do we do with the business judgment rule?
I think Drew is basically correct that this was a catastrophe on many levels. Most disasters occur that way. Still, I think the ratings agencies play a more central role. The big three rating agencies are the only ones approved to grant official ratings. Perhaps we need to open this up to more competition. Maybe we should have more transparency in the ratings process and be better able to compare their success at rating.
Megan goes out of her way to defend the bankers IMHO. Yes, many lost money. Many are still wealthy after losing all that money. Many lived like kings for a while. Fable of the Corn King?
Steve
Limited purpose banking and allow private firms to enter the ratings market. Keep the one’s that are in there, but let in outside firms to compete directly with them.
Sam,
More accurately, the inumerate were bewitched, and as Drew noted the mathematics are incomplete. How else are you going to gauge something extremely complicated? Ouiji board? The Math is utterly necessary. Here is the kernel of the article.
That is the key right there. Every model is an abstraction form reality…a simplification of reality. As such it has its uses, but it is, by definition, limited. Failure to appreciate those limitations can lead you asstray. In this case the failures were devestating.
Even the article botches it to some extent. Instead of blaming the people using the model they blame the model. Bad model, bad! But the model is like a computer it does what its users tell it to do. And just like how a computer can be used for good (guiding airplanes safely through our skies, regulating transportation systems, etc.) they can also be used for bad, identity theft, distributing child porn, and viruses.
It wasn’t that the mathematics were bewitching…at least not to those who understand math, but it was bewitching to those who don’t understand math. They see something that can “plug-n-chug” as I like to call it. No thinking, just pop in the numbers out comes the “solution” and off you go, hundreds of millions of dollars move and yet you don’t really know what you are doing. Its one of my pet peeves with canned statistical software packages for years. It lets people with no understanding or formal training do stuff, sometimes idiotic stuff, that looks complicated and if it gives what looks like the right (or even better the desired) answer then why worry.
Uhhhmm, the understanding the quants had of the model is irrelevant if the above is true. Even if they did know the limitations, it might be that their managers would ignore them. Ignoring the geeky guy sitting 3 cubicles away is often easy, especially if you don’t understand everyother word coming out of his mouth (wtf is a martingale anyways?!!? I don’t know, but look, after plugging-n-chugging we could make tons of money.).
The mathematics will likely always be incomplete. As such any finance model should be taken with a grain of salt at the least. The model should be checked against real world data, simulations should be done, and people should listen to the geek 3 cubicles down who does know wtf a martingale is. Of course, this will likley not happen, but hey you never know.